Performance of a deep neural network in teledermatology: a single-centre prospective diagnostic study.

BACKGROUND: The use of artificial intelligence (AI) algorithms for the diagnosis of skin diseases has shown promise in experimental settings but has not been yet tested in real-life conditions. OBJECTIVE: To assess the diagnostic performance and potential clinical utility of a 174-multiclass AI algorithm in a real-life telemedicine setting. METHODS: Prospective, diagnostic accuracy study including consecutive patients who submitted images for teledermatology evaluation. The treating dermatologist chose a single image to upload to a web application during teleconsultation. A follow-up reader study including nine healthcare providers (3 dermatologists, 3 dermatology residents and 3 general practitioners) was performed. RESULTS: A total of 340 cases from 281 patients met study inclusion criteria. The mean (SD) age of patients was 33.7 (17.5) years; 63% (n = 177) were female. Exposure to the AI algorithm results was considered useful in 11.8% of visits (n = 40) and the teledermatologist correctly modified the real-time diagnosis in 0.6% (n = 2) of cases. The overall top-1 accuracy of the algorithm (41.2%) was lower than that of the dermatologists (60.1%), residents (57.8%) and general practitioners (49.3%) (all comparisons P < 0.05, in the reader study). When the analysis was limited to the diagnoses on which the algorithm had been explicitly trained, the balanced top-1 accuracy of the algorithm (47.6%) was comparable to the dermatologists (49.7%) and residents (47.7%) but superior to the general practitioners (39.7%; P = 0.049). Algorithm performance was associated with patient skin type and image quality. CONCLUSIONS: A 174-disease class AI algorithm appears to be a promising tool in the triage and evaluation of lesions with patient-taken photographs via telemedicine.

Immunohistochemical expression of vitamin D receptor in melanocytic naevi and cutaneous melanoma: a case-control study.

BACKGROUND: Vitamin D deficiency is associated with higher risk of cancer, possibly due to its antiproliferative, antiangiogenic, proapoptotic, cell-differentiating and anti-invasive effects. The anticarcinogenic role of vitamin D in melanoma is still a matter of debate. Loss of nuclear and cytoplasmic vitamin D receptor (VDR) expression in melanoma cells has been reported. OBJECTIVES: To analyse VDR immunohistochemical expression in benign dermal naevi (DN) and malignant melanoma (MM). METHODS: A case-control study evaluated nuclear and cytoplasmic VDR immunohistochemical staining in 54 DN and 55 MM tissue samples. RESULTS: There was significantly higher cytoplasmic VDR positivity in DN compared with MM (59% vs. 16%, P < 0·001). The mean VDR cytoplasmic expression was also higher in DN vs. MM (P < 0·001). No differences in nuclear VDR positivity were observed between groups, but mean nuclear VDR expression was significantly lower in DN vs. MM (P = 0·02). The loss of cytoplasmic VDR in MM was associated with Clark level, tumour staging and American Joint Committee on Cancer pTNM staging (P=0·004, 0·009 and 0·02, respectively). CONCLUSIONS: Alterations in VDR expression and localization are found in MM compared with DN. Loss of cytoplasmic VDR was associated with melanoma tumour size, suggesting that loss of cytoplasmic VDR may be a prognostic factor.

Taxanes-induced cutaneous eruption: another histopathologic mimicker of malignancy.

BACKGROUND: Paclitaxel and docetaxel are antineoplastic drugs that bind the microtubules, producing the arrest of mitoses, which may be seen histopathologically. These histopathologic changes may simulate an intraepidermal keratinocytic malignant neoplasm, and an accurate diagnosis may be only established by clinicopathological correlation. OBJECTIVES: We report six cases of cutaneous eruptions by taxanes in which a striking cytotoxic effect was evident histopathologically. METHODS: Cutaneous biopsies were obtained in each patient. RESULTS: Atypical starburst-like or ring-like mitoses and dyskeratosis on basal and suprabasal layers of the epidermis. Areas of squamous syringometaplasia were also seen in one case. DISCUSSION: These findings were interpreted as expression of mitotic arrest due to taxanes. Similar changes have been described in association with other chemotherapeutic drugs such as vincristine, podophyllin and its derivative etoposide; colchicine, busulfan and maytansine, but cases like ours due to taxanes are exceptional or under-reported. CONCLUSION: Dermatopathologists should be aware of these effects in order to interpret carefully cutaneous biopsy specimens of patients receiving taxanes.

Vitamin D axis and its role in skin carcinogenesis: a comprehensive review

Vitamin D (VD) is a secosteroid hormone that is mainly synthesized in the skin upon exposure to UVB radiation. VD is widely known for its role in calcium metabolism; however, multiple endocrine, paracrine and autocrine functions of VD have been described, including a prominent role on carcinogenesis. In recent years, multiple associations between VD deficiency and different types of cancer have been described, supported by evidence of anti-proliferative, anti-angiogenic, pro-apoptotic, cell-differentiating and anti-invasive effects of this hormone. An immunomodulatory role of VD associated to cancer microenvironment has also been suggested. Regarding skin cancer, it has been shown that VD inhibits tumor development in basal cell carcinoma, squamous cell carcinoma, and melanoma in vitro. Some studies have suggested that lower VD levels may be a risk factor for skin cancer, while others have shown the opposite; there is also preliminary evidence on the role of VD supplementation for the prevention of melanoma in vivo. In this review, we explore the mechanisms of VD effects on carcinogenesis and the available scientific evidence of the interplay between VD and the genesis of both non-melanoma and melanoma skin cancer. (AU)